Research On Monitoring Method Of Structural Strain And Vibration Response Based On Smartphone

Structural health monitoring(SHM)in civil engineering is very significant for ensuring the safety of people’s lives and property.The large-scale application and popularization of the SHM system is the future development trend.The smartphone-based health monitoring system has the advantages of low cost,simple,and portable.It is expected to achieve rapid acquisition of structural response parameters,thereby achieving a preliminary assessment of structural safety,and has been gradually received the attention of scholars in recent years.At present,the appropriate high-precision measurement methods are still lacking,in terms of monitoring the strain response that can reflect the state of structural members.Meanwhile,the technologies of monitoring structural acceleration and displacement response using smartphones are relatively mature and are applied to structural specimens.However,there has been no research on the vibration response monitoring method of three-dimensional(3D)building structures under the action of earthquakes.To address this issue,this paper focuses on the study of the structural strain and vibration response monitoring methods using smartphones.Firstly,a micro image strain sensing method based on the smartphone is proposed to measure the surface strain of structures.Secondly,the method of monitoring the acceleration and displacement response of 3D building structures under the action of earthquakes based on smartphones was studied and the damage of the structure was evaluated.Finally,combining the monitoring methods of the above three parameters,a new type of SHM framework for frame structures is proposed,which further improves the smartphone-based structural health monitoring system.The main research of this dissertation comprises five aspects as follows:(1)A smartphone-based micro image strain sensing method is proposed.The method first uses a smartphone and a portable mobile phone microscope to capture micron-level microimages,and then uses the speeded-up robust feature algorithm and the M-estimated sample consensus algorithm to detect the small displacement of feature points in the image.Finally,the average strain on the surface of a structure at the measuring distance is obtained.Experiments were performed using an unpacked optical fiber as the experimental carrier.The optical fiber with a fiber grating sensor(FBG)is in series to simulate the process of strain generation on the structure surface to verify the feasibility of the method.The results show that using smartphones with different pixels to measure the average strain of different lengths of fibers is in good agreement with the data measured by the FBG,showing high accuracy.(2)A two-point tracking method is proposed to correct the zero-point drift of the microsimage strain sensing method caused by the drift of the smartphone lens focus.To analyze the accuracy and repeatability of the modified strain sensing method in measuring the static strain,a repeatable fiber stretching experiment was performed.Meanwhile,to verify the ability to measure the dynamic strain,a video recording of the fiber during the stretching process was recorded to obtain the dynamic strain of the fiber.The results show that the modified strain sensing method has good accuracy and repeatability,and is capable of measuring the dynamic strain of optical fibers and discriminating the high-frequency component of the strain.(3)A piston sensor is proposed for use with the micro-image strain sensing method.The sensor has a built-in fixed-size circle,which enables automatic calibration of the pixel size in the micro-image.The image barrel distortion due to microscope lens distortion is corrected using the Harris corner point detection algorithm and Gaussian fitting to refine the micro-image strain sensing(MISS)method.To verify the accuracy and stability of the MISS method paired with a piston sensor,experiments were conducted and compared with the results of FBG measurements.The results showed that the MISS method combined with the sensor to measure the static and dynamic strain of the structure has high accuracy and stability.Meanwhile,the sensor is less affected by temperature changes,and the temperature compensation of the sensor can be realized by the empirical formula.(4)A new micro-image strain sensing method that enables different smartphones to to measure structural strain is proposed,called the MISS Ⅱ method.The method uses the arcsupport line segments(ASLSs)circle detection algorithm to identify the moving distance of the center of the moving circle within the new sensor relative to the center of the reference circle to obtain the average strain of structures.The outliers caused by inaccurate identification of circle coordinates are corrected using Hampel filters.To verify the accuracy and stability of the MISS Ⅱ method paired with the sensor to measure the strain of structure,experiments were conducted using two different smartphones and compared with the results of FBG measurements.The results show that the strain measured by the MISS Ⅱ method combined with the sensor has good accuracy and stability.Meanwhile,smartphone software was developed based on the Android platform and cloud server,which can link different smartphones and run the processing through cloud computing to achieve crowdsourcing measurement of structural strain.(5)Aiming at the monitoring of structural acceleration and displacement response parameters,and starting from the study of smartphone measurement of vibration response of 3D building structures,a 3D frame structure model was made and subjected to a shaking table test.Using each bay as the substructure,the accuracy of acceleration and displacement response monitoring by smartphones was investigated by arranging smartphones at the corresponding positions of the substructure.The damage of the structure was evaluated by decomposing the acceleration response data monitored by smartphones using wavelet packets and quantifying the energy change of acceleration response at each node before and after the damage of structure using relative wavelet entropy.(6)A smartphone-based frame structure health monitoring framework is proposed based on the previous studies.The framework is divided into two parts:strain monitoring cloud platform and structural vibration monitoring.In combination with these two parts,the function of monitoring the strain response of force members and the acceleration and displacement response of the structure can be realized.By analyzing these three kinds of data and combining component deflection-deformation curve,structural inter-story displacement angle,and relative wavelet entropy,the damages of local components,substructures,and overall structure are evaluated.This work supplements the framework of the smartphone-based structural health monitoring system.